The three myths of plain English drafting

Daniel Greenberg1 Introduction It is an enormous privilege to have been invited to contribute to this collection of articles in richly-deserved honour of Duncan Berry. I have chosen to write about plain English drafting both in the knowledge that it is a subject dear to Duncan’s heart and also in the pleasurable expectation that he and everybody else will disagree with what I have to say about the subject.

The three myths There are three widely-held myths about the use of plain English in legislative drafting: that it is a modern idea; that readers like it; and that it works. The reality is: that plain English as a dogma is as old as the hills; that nobody who really ought to matter seems to like it much; and that it doesn’t work.

The myth of modernity In every new generation of legislative drafters there arise a number of enthusiasts who are excited to discover that it is possible to draft legislation in more “ordinary” language than is generally in vogue in their jurisdiction at that time. Generally, they fasten upon a stream of criticism from Parliamentarians or others of the complexity of legislation and, either on their own or in partnership with people who have a vested interest in evangelising on the subject of clarity in legislation, they embark upon a campaign to bring light to the heathens. In fact, however, any heathen who has been properly trained in legislative drafting will already have been exposed to the concept of plain language, not as a fad or luxury, but simply as part of the core tools of good drafting. Even a work as venerable and apparently dated as Sir Alison Russell’s Legislative Drafting and Forms2 includes concise and trenchant advice to draft in as simple and clear a style as possible. All the

1

Parliamentary Counsel, Office of the Parliamentary Counsel (UK), 1991-2010; Parliamentary Counsel, Berwin Leighton Paisner LLP, 2010→; Editor, Craies on Legislation, Stroud’s Judicial Dictionary, and Jowitt’s Dictionary of English Law; General Editor, Annotated Statutes, Westlaw UK. 2 London, 1920.

Page 103

The three myths of plain English drafting greatest exponents of the art of legislative drafting, notably Ilbert, Ram and Fiennes, have urged the importance of drafting clearly and concisely, and have consistently practised what they preached.

Sentence length and the theory of relativity So there is nothing new about the concept of plain language drafting; nor is there anything new about any of the techniques presently espoused as keys to achieving a plain drafting style. In criticising the language of legislation, for example, much play is made of the length of particular sentences. It is certainly true that many impenetrable laws can be greatly improved simply by being broken up into a number of shorter propositions. But again, there is nothing new in that: the best exponents of the art of legislative drafting have always also been those who have drafted in a relatively staccato style. In the last sentence, however, “relatively” is the key word. There is no objective gold standard of plain language either as to sentence-length or as to anything else: the standard is set by the ebb and flow of fashion in the natural use of language outside the legislative context. It is noticeable that the tolerance of English readers generally for sentence length has diminished greatly over the years. One hundred years ago it was not unreasonable to expect an average reader of fiction to navigate his or her way through a sentence running to between one hundred and two hundred words, and the most widely read writers of popular fiction did precisely that. By the same token, therefore, there would have been nothing undesirable about a legislative drafter of that time using a one hundred word sentence: it could still have qualified as “plain English”. Now that general tolerance for sentence-length has changed, so too must the general standards of legislative drafting. It is not, however, that we have discovered a new technique of writing in short sentences: it is simply that what amounts to short or long by normal literary standards has changed.

The myth of popularity One of the many problems of legislative drafters is that we spend too much time talking to other legislative drafters, or other people with particular and peculiar obsessions with the language and preparation of legislation. We and they are all enthusiastic, not to say sometimes obsessive, about whatever we think we have discovered as innovative methods of plain language drafting. We draw support from certain campaign groups that have also chosen to make their reputation, not to mention in some cases their living, by espousing the cause of plain language. And taking the two together we satisfy ourselves that the quest for plain language drafting is one that is both in the public interest and thoroughly supported by all. It may be salutary to remind ourselves, therefore, that for most people this quest is rather less urgent and important than we like to think. Worse than that, while for some the question is simply a bit of a yawn, for others the trend towards plain language drafting is actually unwelcome. I have suspected this for some time, but received the most concrete evidence of it when the Department for Environment, Food and Rural Affairs published in 2009 a draft Bill of mine about Floods and Water Management3. The consultation paper accompanying the draft Bill included, as an innovation, a short series of questions about the style of the draft, inviting readers to comment on whether the language used was appropriate (see Appendix). I had tried to draft the Bill in a clear and simple style, and to use plain language so far as possible, and this was noticed and approved by the more “legal” of readers, in particular the Law Reform Committee of the General Council of the Bar.

3

http://www.official-documents.gov.uk/document/cm75/7582/7582.pdf

Page 104

The three myths of plain English drafting So those readers who might have been expected to be able to cope with a more complex and technical style enjoyed the simplicity of the draft. But those for whose comfort and convenience the style was particularly designed were apathetic at best and hostile at worst. In particular, in various places I had used “thinks” and “thing” in preference to words like “considers” and “substance”, congratulating myself on avoiding archaic or legal language in favour of the kind of language that one would use naturally in the course of casual conversation. I told myself that I had lost nothing in meaning but gained in stylistic clarity and simplicity. Some of my readers, however, disliked the product for precisely the reason I had thought it praiseworthy: one consultee actually felt that more “legal” language would have been more appropriate, and the same feeling was voiced, less expressly, by a number of others. Analysing the responses informally by class, the further the reader was away from the legal profession the less he or she appeared to appreciate my well-meaning attempts to use simple language.

People like judges to look like judges, and law to read like law The reality therefore appears to be that people like their law a little pompous, an element of pomposity being generally associated with authority. People like to see judges in wigs and gowns, and the pressures to make judges look more like “ordinary” people tend to come from people in or close to the legal profession who specialise in thinking that they know what “ordinary” people like. “The people”, whoever they may be, like to think of the law as something a little distanced from them, and a spot of grandeur helps to preserve distance and imbue authority. The same, apparently, applies to the language of the law. Lay-readers appear to relish a certain distance and weight in legislation, being uncomfortable with the idea that it should exactly reflect the language that they would choose for ordinary conversation. The law is not conversation, and it appears that attempts to present it as if it were do not find favour with those whom they are designed to please. Nor is it only lay-readers who dislike the “dumbing-down” of legal language. For several years I did my best to replace what I saw as older and more pompous expressions such as “considers” and “is of the opinion” with the simple “thinks”. Time and time again I encountered resistance from Parliamentarians who thought that “considers” sounded better – by which they appeared to mean simply that it sounded less like the kind of language one might use when talking to a friend in the pub. Sometimes I won, because I had a Minister on my side who wanted to modernise legislative language and the political stakes were relatively low; sometimes I was forced into ignominious retreat by way of amendment. I comforted myself with the thought that the more often I won the easier it would be to cite precedent to support the next attempt; but I wonder now whether the game was worth the candle or, to put it differently, whether I should have been playing that particular modernising game at all if it was pleasing neither the lay-readers nor their Parliamentary representatives.

The myth of efficacy The really dangerous myth about the use of plain language in legislative drafting is the myth that it works. The reality is that for a variety of reasons legislation can never be presented in a fashion that makes it readily intelligible to a lay-reader, and the more it uses plain English the more danger there is that the lay-reader will be misled into believing that he or she has understood more of it than is actually the case. The fundamental difficulty in writing legislation is that we are doing two things simultaneously and within a single document: we are making the law, and we are also communicating it. We have to fulfil the latter function using English (or another spoken language), and an English that as closely as possible matches how

Page 105

The three myths of plain English drafting people communicate in the real world. But the first part of the process can only be achieved by engaging the concepts of which the law is composed; concepts which are technical and complicated and for many of which demotic English does not provide an exact equivalent. When a legal concept has no near equivalent in ordinary English, the problem is bad enough. So, for example, exercises in translating existing Acts into plain English routinely go somewhat as follows— 1. They take all the provisions which are pretty much intelligible anyway and turn them into a form of language which is either more or less elegant and pleasing than the original, depending on what you like. 2. As I say above, in that part of the process they will doubtless please themselves and their campaigning colleagues, but it is a moot point whether they will please as many of those whom they claim to be championing as they think. 3. In the process of beautifying the language they will probably destroy one or two subtleties, and perhaps make one or two unintended changes of minor policy without realising it. 4. Then after a subsection or two they come across a provision that is entirely engaging a technical legal concept and has no demotic near-equivalent – such as a provision to the effect that “A person may not be required under this section to answer a question that he or she could not be forced to answer in the course of civil proceedings”; which may be designed to apply the rules of legal professional privilege, the privilege against self-incrimination, or both. 5. Unable to render the concept into “plain English”, the translator simply repeats it verbatim, and perhaps adds an acknowledgment that there are of course limits to what can be done in the way of plain English. There are indeed very significant limits on what can be done by way of translating technical legal concepts into ordinary English, and they are sometimes of such an impact on the material as to render the entire process more or less pointless: what is gained by an “ordinary” reader being able to understand (or think that he or she has understood) subsections (1) to (6) easily, if a major restriction on the effect of those subsections is couched in such technical form that they cannot be sure whether or not the clause as a whole applies to them without obtaining expert professional advice? Of course, there is no alternative to using technical legal concepts in this way. We cannot write out a fiftypage treatise on the law of evidence every time we impose a requirement to give information; and even if we could, it would be out of date as soon as written. The only way of making the law in an effective form is to engage a long-standing and constantly developing concept such as legal professional privilege. But although that is the right answer for process number 1 – making the law – it does not lend itself to any easy solution for process number 2 – communicating the law. Oddly enough, the problem which we face when there is no near equivalent in ordinary English to a particular legal concept is less severe than the problem that we often face when there is. A good example of this is the use of the word “person”, which appears many times in most Acts of Parliament. A non-expert reader who comes across the word “person” recognises it as a slightly formal or stilted, but still extant, word for human being. A legally qualified reader may congratulate himself or herself on recognising it as being used as a method of engaging the technical legal concept of something with legal personality. But they are both wrong – because the Interpretation Act 1978 provides that as used in legislation the concept includes both natural and unnatural persons, including bodies such as unincorporated associations that do not have legal personality. The result is that any reader unfamiliar with the technicalities of the 1978 Act is likely to think, wrongly, that they have understood the provision, whereas in fact they have been misled as a result of the word used having a similar, but not equivalent, use in “ordinary” English. Page 106

The three myths of plain English drafting Attempts to make this easier for readers frequently make the problem worse. A number of drafters of recent Acts, for example, have repeated the Interpretation Act 1978 definition as an express definition within another Act. But that creates a puzzle for readers of other similar Acts that do not repeat the definition, to try to decide whether those Acts that do repeat it are trying to override a contrary indication, and if so whether there is a similar, but non-overridden, contrary indication in those Acts that do not repeat it. To take another example, a number of drafters have recently started to use “people” as a plural – the problem being that it is unclear whether it is merely being used as a spiritedly fashionable attempt at a plain English equivalent of “persons”, or whether it is being used as a word that is naturally apt only for individuals, thereby amounting to a contrary indication to the inclusions in the Interpretation Act.

Conclusion If plain language drafting doesn’t work and our readers don’t want it, should we abandon it and go back to two hundred-word sentences full of hereuntobefores and other antediluvian delicacies? No. The drafter who is punctilious about getting the law “right”, in the sense of producing a draft that gives effect to the policy, but doesn’t care how long it takes the reader to satisfy himself or herself to reach the same conclusion, is making two mistakes. First, he or she is performing only one of the two parts of our job – making the law, but not communicating it effectively. Worse, however, if the drafter is satisfied that the law “works”, but nobody else finds it easy to discover whether they agree, then the law doesn’t work: if the people at whom a legal duty is aimed cannot work out what they are meant to be doing, the law is not effective in practice, however perfect it may be in a theory that exists only in the drafter’s mind. So we should continue the search for ways of presenting law in a manner that makes it as easy as possible for the most likely target audience or audiences of each Act to understand what it is intended to mean, with as little recourse to extraneous explanatory material and advice, or to the courts, as possible. But we need to conduct the search in a scientific way. In particular, we must resist the temptation to assume that as drafters we are in a position to know how easy our different target audiences will find it to assimilate and understand different stylistic approaches. There is no evidence that we are adequate predictors of what our readers like. In all other respects we demand solid evidence to underpin our work on legislation, and modernisation of style should be no different. Nor should we take our evidence on the matter from selfappointed campaign groups, unless there is a specific reason to accept a particular group as likely to be in a position to speak authoritatively for the intended target audience of a particular piece of legislation. All of which points towards readers panels, drawn from “real users” of legislation, and representing different target audiences. So, for example, a Finance Bill provision about the taxation of oil futures should be tested on specialist industry lawyers or accountants (preferably both) not purely for discussion of the policy and substance but also for confirmation that the legislative language engages the appropriate technical terms and jargon to reflect the language of the context within which the legislation will fall to be applied. There is no merit in using jargon for the sake of it; but nor is there merit in avoiding the use of jargon which is generally used and understood within a trade or profession, if the only alternative is to attempt to construct the same concept out

Page 107

The three myths of plain English drafting of “ordinary” words, an attempt which at its best will be unhelpfully inelegant and at its probable worst will simply be wrong. A piece of legislation about regulating dogs in parks, however, should be tested on a panel composed of citizens with no particular expertise in law, but simply with a public-spirited willingness to assist us in framing legislation about everyday matters in a clear and simple way. In particular, it is important to use readers panels composed of people who have no financial or other vested interest in promoting a particular set of “plain English” approaches. As we engage with our readers panels, we should also take the opportunity to explain and explore the limits of what we can do to make legislation accessible by modernising our language. As well as working on different kinds of explanatory materials and tools, we can also try to educate readers so that they are less likely to be misled into thinking that they have understood more than is actually the case. Good legislative drafting is about a continual search for improvement, keeping step with changing fashions in the use and assimilation of language in other contexts. That search is not a new one, but from time to time it can take new turns by harnessing political or social emerging trends. The modern mania for consultation, although often not particularly useful, can be turned to good use in the context of legislative drafting if we take the opportunity to engage our target audiences’ experience and knowledge to inform and guide our attempts to help them, and consequently to enhance the rule of law by making legislation easier to understand.

Appendix Drafting questions that accompanied Floods and Water Management Bill: 1. How far, in general, would you say that the draft legislation is written in a reasonably clear style that is likely to be understood by readers? 2. In general, do you think the individual clauses are too long, too short or about the right length? How far is their overall order in the draft legislation reasonably logical and easy to follow? 3. In general, do you think the individual sentences in the draft are too long, too short or about the right length and is their structure too complex, too simple or about right? 4. Please give examples of anything in the style of the draft legislation that you particularly liked or disliked. Please also give your reasons. 5. Please give examples of provisions that you thought helpfully simple or well expressed or ones that could be made simpler or otherwise improved. Please also give your reasons. 6. Are there any drafting techniques (such as cross-references to other provisions of the draft legislation) that you would like to see used more or less? 7. Please suggest any improvements to the way in which legislation is drafted that you think would make it easier to understand and apply.

Page 108